Pharmas don’t publish all their results.
Thomas Edison said once “I have not failed. I have found 1,000 ways that won’t work.” Fast forward to today, where 90 percent of drug trials end up as failures. The difference between Edison and today, is that you may never know the details of those 90 percent. Why? Because drug companies are under no obligation to publish their trial results, and are particularly reluctant to publish negative results.
Why should drug companies publish their trial results?
Obviously, drug trial data is submitted to regulatory bodies to make the scientific case for approving a new drug. But sometimes data isn’t made public, especially if the results end up in failure. The failure could come after regulatory submission, or may persuade researchers and pharmaceutical companies to not pursue a drug any further. The problem with this is, we never get a full sense of how a chemical compound worked in the body without as much data as possible. In addition, Ben Goldacre of BadScience.net (and a former science columnist for The Guardian) maintains that such missing data could falsely persuade doctors to prescribe new drugs when an older, less costly alternative will do the trick.
How much data is missing?
The British National Health Service estimates that about half of all drug trials were never published in an academic journal. In addition, negative results are twice as likely to be ignored as positive results. And it doesn’t matter whether the trial is large or small, in an early phase or a later one, or performed by private industry or a public institution; only half see the academic light of day. This creates an obvious bias in studying the effectiveness of drugs, and how the body reacts to them.
Whose fault is this?
A lot of people’s, apparently. Both the US FDA and European Medical Agency are supposed to require clinical trial results to be posted publicly, and published. But no auditing process exists, and so far, relatively few trials are actually posted, Goldacre says. Universities aren’t enforcing (or writing) contracts that force results to be published, no matter what the results. Various ethics committees don’t require publication of results, medical membership associations haven’t stepped up to require more publication.
What’s the risk of not knowing these results?
Medicines react to the body in different ways, and many times, trials fail not because the drug candidate was an outright failure, but the dosage was wrong, or it was addressing a large group of people instead of a subgroup that would respond positively (this latter issue happens a lot in cancer trials). Sometimes, results can prevent disasters or even save lives; the side effects of Vioxx and other similar drugs were found by doctors and academic researchers, not by regulators approving the drugs (and since, those drugs have been taken off the market).
What to do?
Earlier this year, Goldacre testified to the British House of Commons, and that country’s National Audit Office is also looking at the problem. A website, www.alltrials.net, is advocating for publication of all trial results. For now, pressuring government and making more people aware of the problem are big first steps.
Sources: www.badscience.net, www.alltrials.net
Ross JS, Mulvey GK, Hines EM, Nissen SE, & Krumholz HM (2009). Trial publication after registration in ClinicalTrials.Gov: a cross-sectional analysis. PLoS medicine, 6 (9) PMID: 19901971
Ross JS, Tse T, Zarin DA, Xu H, Zhou L, & Krumholz HM (2012). Publication of NIH funded trials registered in ClinicalTrials.gov: cross sectional analysis. BMJ (Clinical research ed.), 344 PMID: 22214755
goldacre bad pharma, pharma trials, KEYWORD